Early initiation of drug abuse is associated with increased likelihood of later dependence and a decreased probability of quitting in adulthood. These observations are supported by results of animal studies demonstrating that exposure to CNS-active drugs around the time of adolescence causes behavioral changes related to drug consumption that persist to adulthood. We have obtained similar results investigating effects of cannabinoids (drugs that act like marijuana) administered during periadolescence on the vocal development of a songbird, the zebra finch, an animal that learns a form of vocal communication during periods of late-postnatal development. We have found that cannabinoids alter zebra finch vocal development (butnot already-learned adult song) and that the neural circuit that regulates this learning process contains a strikingly distinct and dense expression of cannabinoid receptors. Thus, the songbird brain provides a unique model to examine the effects of cannabinoids on neural and behavioral development. This new project will identify physiological changes responsible for cannabinoid-altered vocal learning. Because late-postnatal CMS development is characterized by similar activity-dependent processes of synaptic refinement and fiber tract maturation in both songbirds and mammalian species, what we learn about the neural mechanisms of cannabinoid-altered sensory-motor vocal learning will likely prove relevant to the problem of human marijuana abuse that typically begins during adolescence.